Abstract
Campylobacter is a major foodborne pathogen and its antimicrobial resistance (AMR) has been described worldwide. The main objective of this study was to determine the occurrence and AMR of Campylobacter spp. isolated from broilers throughout the supply chain in Valencia, Spain. A total of 483 samples were included in the analysis: 430 from the slaughterhouse (chicken carcass and neck skin) and 53 from the point of sale (retail broiler and packaging). Taking into account the origin of the sample, the prevalence of Campylobacter spp. was 19% in carcass, 28.2% in neck skin, 36.7% in retail broiler, and 80% in packaging isolates. The prevalence of different species in the analyzed samples was 21.1% and 4.8% for Campylobacter jejuni and Campylobacter coli, respectively. AMR profiling of 125 Campylobacter isolates revealed that 122 (97.6%) of the isolates were resistant to one or more antimicrobials. C. jejuni samples presented high resistance to nalidixic acid and ciprofloxacin, 96.1% and 90.2% respectively, whereas C. coli showed 87% of resistance to both antimicrobials. Both species were resistant to tetracycline (C. jejuni 84.3% and C. coli 60.9%) and 26.1% of C. coli was resistant to streptomycin. These results showed no significant difference in the frequency of AMR (p ≥ 0.05) among isolates originated from different points in the food-processing chain at slaughterhouses and retail establishments. In contrast, three main patterns were detected: quinolone–tetracycline (64%), quinolone-only (17.6%), and quinolone–tetracycline–aminoglycosides (8%). Additionally, 12.8% of the isolates presented multidrug resistance, with significantly higher levels detected among C. coli (30.4%) isolates compared with C. jejuni (8.8%) and all the three strains were resistant to all six antibiotics tested. Therefore, these results indicate that broilers could be a source of antimicrobial-resistant Campylobacter in humans and consequently pose a risk to public health.
Introduction
Campylobacteriosis is currently the most commonly reported zoonotic bacterial foodborne gastroenteritis worldwide (Igwaran and Okoh, 2019), and has been so over the last decades in different parts of the world (Heimesaat et al, 2021), including developed and developing countries (Dixon et al, 2022; Igwaran and Okoh, 2019). In 2020, the number of confirmed cases of human campylobacteriosis totaled 120,946, corresponding to a European Union (EU) notification rate of 40.3 per 100,000 population. Of these, 88.1% were Campylobacter jejuni, and 10.6% Campylobacter coli (EFSA-ECDC, 2021).
Campylobacter is an enteric and zoonotic organism well adapted to the intestinal environment of various food animal species and is highly prevalent in ruminants, swine, and poultry (Shange et al, 2019). The reported occurrence of Campylobacter in the main food categories in 2020 within the EU shows that meat and meat products were the most contaminated food categories, followed by milk and milk products, fruit, vegetables, and juices. The percentage of Campylobacter-positive units was the highest for fresh meat from broilers (30.1%), followed by other fresh meat (25.1%), and meat from turkeys (21%). The ratio of fresh meat from pigs and bovines remains relatively low at 3.7% and 0.4%, respectively (EFSA-ECDC, 2021). These data support the fact that poultry, particularly broiler chickens, are the main source for human infection with Campylobacter.
Water and feed contamination is the most notable source of contamination during primary production. Cross-contamination of broiler carcasses, skin, and meat occurs during slaughter, dressing, and processing through machinery, work surfaces, water, and air, partially due to leakage of contaminated feces from visceral rupture (Mota-Gutierrez et al, 2022). Epidemiological studies have shown that two important pathways exist for human exposure to Campylobacter owing to contaminated poultry: eating undercooked poultry meat and crosscontamination events (Santos-Ferreira et al, 2021).
Most Campylobacter infections are self-limiting and do not require antimicrobial treatment. However, severe and prolonged cases of campylobacteriosis and infections in immunocompromised, vulnerable populations and children may require antimicrobial therapy. In such cases, quinolones and macrolides are the drugs of choice (Dai et al, 2020; Narvaez-Bravo et al, 2017). Tetracycline can also be used as an alternative treatment (Alaboudi et al, 2020). However, when complications arise, antibiotic therapy is required at a time when resistance to antibiotics is increasing worldwide. When these drugs are ineffective, systemic administration of aminoglycosides is the only option left (Aleksić et al, 2021; Lopez-Chavarrias et al 2021; WHO, 2017).
Since the beginning of the 1950s, antimicrobial agents have been included in animal feeds to improve growth and reduce production costs (Dodds, 2017). However, the increasing use of antimicrobials in animal husbandry and aquaculture to treat diseases, prevent infections, and promote livestock growth and feed efficiency has become problematic (Cheng et al, 2019; Murphy et al, 2017). The use of antimicrobials in an untargeted manner, at subtherapeutic doses, repeatedly, or for inappropriate periods of time in food-producing animals, can lead to the development of multidrug-resistant (MDR) zoonotic bacterial pathogens, failure of prevention and life-saving treatments, and serious threats to global human health (European Commission, 2015; Pérez-Rodríguez and Mercanoglu Taban, 2019). Antimicrobial resistance (AMR), including MDR, is a global problem that poses a serious threat to human health.
The aim of the present study was to determine the prevalence and AMR of Campylobacter in poultry in Valencia (Spain) during 2016–2017, as a consequence of the “Strategic and action plan” established by the European Commission on the monitoring and reporting of AMR in zoonotic and commensal bacteria (European Commission, 2013).
Materials and Methods
Sample collection
The Campylobacter isolates were recovered by the Public Health Laboratory of Valencia, between January 2016 and December 2017, in Valencia area, Spain. Chicken carcasses (inner surface) and neck skin sampling for microbiological analysis were taken after cooling from three different industrial slaughterhouses, and was carried out in accordance with the ISO 17604:2015 standard. Retail packaged broiler and packaging were obtained from six supermarkets, and 25 g portions of the thigh and breast were used to study. In addition, to determine the number of sample units that should be collected, Annex I of Regulation No. 2073/2005 of the Commission of November 15, 2005 on the Microbiological Criteria applicable to food products was used as a reference.
Campylobacter isolation and characterization
Campylobacter isolation procedure was performed according to ISO 10272–1:2017 and ISO 10272–2:2017 after enrichment in Bolton Broth (Liofilchem srl., Spain). To 25 g of each sample, 225 mL of Bolton Broth was added (1/10), samples were stomached for 60 s, and incubated in a microaerobic atmosphere (5% O2, 85% N2, and 10% CO2) using a generator sachet (bioMérieux, Spain) for 4–6 h at 37°C ± 1°C, followed by incubation for 44 ± 4 h at 41.5°C ± 1°C. Each sample was spread on modified Charcoal Cefoperazone Deoxycholate Agar (VWR Chemicals, Spain) and incubated at 41.5°C ± 1°C microaerobically for 24–48 h.
Presumptive Campylobacter colonies were subcultured on Columbia Blood Agar (VWR Chemicals) for 24–48 h at 41.5°C ± 1°C. Typical colonies were examined using dark-field microscopy. In isolates that had typical morphology and motility and for which the oxidase test (Merk, Spain) was positive, gram staining, catalase activity (VWR Chemicals), hippurate hydrolysis (Thermo Scientific™ Hippurate Disc Remel™, Spain) growth at 25°C ± 1°C and 41.5°C ± 1°C in a microaerobic environment, and growth at 37°C ± 1°C in an aerobic atmosphere were tested for identification. The identification of the isolates was performed using the API CAMPY system (bioMerieux, Spain), and confirmed at the Microbiology Laboratory of the University of Valencia, using the automated Thermo Scientific Sensititre™ Identification system (Thermo Fisher Diagnostics), according to the manufacturer's instructions.
Antimicrobial susceptibility testing
The isolates were tested for antimicrobial susceptibility to six antibiotic agents (nalidixic acid [NAL], ciprofloxacin [CIP], erythromycin [ERY], tetracycline [TET], gentamicin [GEN], and streptomycin [STR]) by the broth microdilution method using Sensititre-Campylobacter-EUCAMP2™ microtiter plates containing twofold serial dilutions of the six antibiotic agents, according to the manufacturer's instructions (Thermo Fisher Scientific™, Madrid, Spain). The plates were sealed and incubated at 37°C for 48 h according to the manufacturer's instructions, and they were visually read using Sensititre™ Manual-Viewbox (Thermo Fisher Scientific). C. jejuni and C. coli isolates were categorized as susceptible or resistant to antimicrobials using the epidemiological cutoff values (ECOFF) published by EUCAST (
Statistical analyses
IBM-SPSS-Statistics for Windows, version 25.0 (Released 2017; IBM Corp., Armonk, NY) was used for statistical analyses of the data. For each antibiotic tested, the proportion of resistance found in slaughterhouse and point of sale was compared using the chi-squared test. The level of significance has been set at p < 0.05.
Results
Four hundred eighty-three samples were included in the analysis: 430 from the slaughterhouse (chicken carcass n = 253 and neck skin n = 177) and 53 from the point of sale (retail broiler n = 48 and packaging n = 5). The prevalence of Campylobacter was 25.9% (125/483), and C. jejuni was the most frequent species in both types of samples. Taking into account the origin of the sample, 19% from carcasses, 28.2% from neck skin, 36.7% from retail broiler, and 80% from packaging were positive for Campylobacter (Table 1).
Prevalence of Campylobacter Collected from Broilers at Slaughterhouse and Point of Sale
n, total positive isolates.
The Campylobacter spp. isolates displayed a remarkable resistance to nalidixic acid (90–100%) and to ciprofloxacin (78–100%) similar for both species, and results are similar when they are referred to the species (Table 2). Isolates displayed a significant resistance to tetracycline (69.6–100%), and lower to other antimicrobials: erythromycin (2.1–4.3%), gentamicin (2.1–6%), and streptomycin (8.7–14%) (Table 2).
Antimicrobial Resistance of Campylobacter jejuni and Campylobacter coli Isolated from Different Samples
n, total isolates; n +, total resistant isolates.
Overall, according to the data obtained, 87% and 96.1% of C. coli and C. jejuni samples, respectively, demonstrated remarkable resistance to nalidixic acid, similar to ciprofloxacin (87% and 90.2%). Tetracycline resistance was 60.9% for C. coli and 84.3% for C. jejuni and 26.1% of C. coli was resistant to streptomycin. Resistance to the remaining antibiotics was observed to a lesser extent (Table 2). All p-values were >0.05, and no significant differences were found between antibiotic resistance and sample origin.
In addition, 97.6% (n = 122) of the isolates analyzed (n = 125) were resistant to at least one antibiotic. The distribution of resistance patterns is shown in Figure 1. In total, 16 (12.8%) isolates (C. jejuni, n = 9; C. coli, n = 7) were resistant to 3 or more drug classes and were defined as MDR strains, 13 isolates were resistant to 3 classes of antibiotics, and 3 isolates were resistant to all classes.

Distribution of resistance patterns. MDR, multi-drug resistance.
The AMR patterns of the resistant isolates are summarized in Table 3. Overall, the MDR rate was significantly higher in C. coli than C. jejuni 30.4% (7/23) versus 8.8% (9/102). The most common MDR pattern was the combination of quinolones/tetracycline/aminoglycosides, which was observed in 21.7% (5/23) of C. coli and 7.8%. (8/102) of C. jejuni isolates. Two of the C. coli and one of the C. jejuni samples were resistant to all the antimicrobial classes tested, and two of these isolates corresponded to retail broiler samples (Table 3).
Antimicrobial Resistance Profiles of Campylobacter Isolates
CIP, ciprofloxacin; ERY, erythromycin; GEN, gentamicin; n, total resistant isolates; n +, total positive profile isolates; NAL, nalidixic acid; STR, streptomycin; TET, tetracycline.
According to the type of sample, 48 strains isolated from chicken carcasses were resistant to ciprofloxacin (97.9%) and the same to nalidixic acid (Table 2), of which 16.7% presented a resistance profile CIP/NAL, whereas 66.7% also showed resistance to tetracycline (CIP/NAL/TET) (Table 4). Furthermore, eight MDR isolates were found, two of which showed resistance to three antimicrobial classes, seven to quinolones/aminoglycosides/tetracycline, and one to quinolone/macrolides/tetracycline (Table 3).
Antimicrobial Resistance Profiles of Campylobacter Isolated from Slaughterhouse and Point of Sale
Of the 50 strains isolated from chicken neck skin, most were simultaneously resistant to the two quinolones (90%), of which 60% presented a CIP/NAL/TET profile and 16% CIP/NAL profile. Seven of the microorganisms studied were MDR, with two strains showing resistance to all antibiotics, one from C. coli, and the other from C. jejuni (Table 4).
Most of the 23 strains isolated from retail broiler were resistant in parallel to the two quinolones (95.7%), and in 60.1% of them the resistance profile was CIP/NAL/TET and in 26.1% CIP/NAL. One strain (C. coli) showed resistance to all antibiotics analyzed, with only two MDR strains (Table 4).
Finally, in relation to the samples from chicken packaging, the four isolates were simultaneously resistant to the quinolone group and tetracycline. No resistance was observed for erythromycin, gentamicin, or streptomycin. In addition, MDR was not detected (Table 4).
Discussion
The acquisition of MDR is one of the most important public health concerns (EFSA-ECDC, 2018a; Giurazza et al, 2021). Campylobacteriosis is the main zoonosis prevalent in Europe (EFSA-ECDC, 2021), mainly transmitted by meat consumption. In our study, the proportion of Campylobacter-positive samples found (25.9%) (Table 1) was lower than the values reported by European Food Safety Authority (EFSA) from EU Member States in 2016 and 2017 (36.7% and 37.4%, respectively) (EFSA-ECDC, 2018b; EFSA-ECDC, 2017), reaching in Spain 51, 27% (MAPA, 2019). However, comparing data between studies is often difficult because of the different origins of the samples and countries, time periods, methods of sample collection, and laboratory analysis.
In addition, our study observed that C. jejuni was the predominant species isolated. This result was similar to most studies worldwide, in which C. jejuni was the most prevalent species isolated from chickens, followed by C. coli (Kouglenou et al, 2020; Linn et al, 2021), which has been responsible for most gastroenteritis cases during at least the last few years (EFSA-ECDC, 2021; Zbrun et al, 2020).
Furthermore, among the total analyzed products, retail broiler presented the highest prevalence (36.7%) compared with samples from slaughterhouses (19% carcass and 28.2% neck skin) (Table 1). This could be explained by the fact that Campylobacter contamination is strongly influenced by handling after the slaughterhouse, and crosscontamination might have an important role, since it might increase Campylobacter prevalence (García-Sánchez et al, 2018; Soro et al, 2020). These results are in accordance with studies that found the highest occurrence of Campylobacter in retail meat products (Kwon et al, 2021). Rasschaert et al (2020) and Alter and Reich (2021) not only focused on the handling of chickens in the slaughterhouse, but also pointed out loading and transport and their subsequent processing and conservation as critical points.
Moreover, the percentage of contamination inside the outer packaging of retail chickens was 80% in this study. Although there are very few studies found in the literature on this topic, the Food Standards Agency of the United Kingdom examined the levels of contamination by Campylobacter in >3000 samples of freshly chilled whole chickens and their packaging, purchased in large supermarkets, small stores, and butchers. The results indicated that depending on the establishment, Campylobacter contamination of packaging ranged from <1% to 20%, with an average of 7–8% (Public Health England, 2021). In addition, in the study conducted by Harrison et al (2001), Campylobacter was isolated from 3% of external and 34% of whole packaging. Although the percentage of contamination found in our study is higher, possibly due to the low number of samples analyzed, we can conclude that chicken packaging is a vehicle for potential crosscontamination of Campylobacter.
Regardless of the origin of bacteria, 97.6% of the Campylobacter isolates analyzed were resistant to at least one antibiotic. In 2016, the EFSA reported that 72.3% of C. jejuni and 89.5% of C. coli isolated from broilers in the EU were resistant to at least one antibiotic; and high to extremely high proportions of isolates were resistant to ciprofloxacin and tetracyclines. However, EFSA informed significant differences in AMR rates between the EU Member States, being notably lower in northern countries and higher in southern countries, especially Spain (EFSA-ECDC, 2018a). These differences are explained by McCrackin et al (2016) in a systematic review of literature, who could not decide on a causal relationship between use of antimicrobials in agricultural animals and the prevalence of drug-resistant foodborne campylobacteriosis in humans, but concluded that, on the farm, antibiotic selection pressure could increase colonization of animals with drug-resistant Campylobacter spp.
The present study showed a high rate of resistance to quinolones (94.4%). Over the last two decades, there has been a rapid increase in the percentage of resistance to ciprofloxacin worldwide. The EFSA (EFSA-ECDC, 2018a) reported that an alarming situation was found in Spain, as the rate of resistance detected was >88%, which is lower than our result. However, although the level of resistance obtained in this study is comparable to those obtained by other authors in other EU countries (Di Giannatale et al, 2019; Marotta et al, 2019), the results of all EU Member States show a considerably lower rate, arguing that in Europe, the incidence of resistance to quinolones in broilers is highly variable, ranging between 8% and 10% in Finland and Norway and 98% in Latvia (EFSA-ECDC, 2018a).
Some studies have linked resistance to the use of these antimicrobials in animals intended for human consumption, particularly in the poultry industry (Cobo-Díaz et al, 2021; ECDC-EFSA-EMA, 2017a; Roth et al, 2019). In countries where the use of quinolones in poultry production is not allowed, such as Denmark, Finland, and Sweden, a few resistant strains of Campylobacter have been isolated from chickens or humans (ECDC-EFSA-EMA, 2017b).
In addition, the high percentage of tetracycline-resistant isolates obtained in this study, in Spain in general and in our study in particular, is probably related to its inappropriate use in poultry production owing to its low cost and easy application to animals (ECDC-EFSA-EMA, 2017a; Gahamanyi et al, 2020; Rivera-Gomis et al, 2021). Moreover, tetracycline resistance has been reported to persist for years even in the absence of antibiotic selection pressure (Luangtongkum et al, 2009).
The incidence of gentamicin, streptomycin, and erythromycin resistance has been very low, in accordance with the data obtained by the EFSA-ECDC (2018a), and it was posited that the low resistance to these antibiotics was due to their rare use in prophylaxis and therapy in the poultry industry. Moreover, almost all isolates obtained in this study that showed resistance to these antimicrobials also exhibited MDR. Lopez-Chavarrias et al (2021) suggested that coresistance to both aminoglycosides and macrolides, as well as MDR to additional antimicrobial classes, could be explained by the possible circulation of resistance genes against different antimicrobial classes.
Furthermore, the prevalence of MDR in C. coli (31.8%) was higher than that in C. jejuni (9%), which is consistent with the results of previous studies (Hull et al, 2021; Linn et al, 2021). In total, 12.8% (16/125) of the isolated Campylobacter strains were classified as MDR with at least three antibiotics belonging to three different families and five MDR profiles; four were observed in the C. jejuni isolates, whereas three patterns were detected in the C. coli isolates. The main MDR profile in Campylobacter spp. isolates was the combination of quinolones, tetracycline, and aminoglycosides (Table 3). Additionally, three isolates were resistant to all six of the tested antibiotics. However, these microorganisms are epidemiologically significant not only because of their resistance to multiple antimicrobial agents, but also because of their ominous prospect of being resistant to almost all or all approved antimicrobial agents, which is a serious concern for human health.
Finally, this study showed that the AMR profiles of the Campylobacter spp. isolates were similar among different kinds of samples (slaughterhouse and point of sale), suggesting that resistance in Campylobacter found in broiler chicken meat sold at retail and packaging sites originated upstream at the farm. These results are compatible with those obtained in a study by Dramé et al (2020) in Canada regarding the level of resistance in isolates among sectors of the broiler chicken supply chain.
In conclusion, a high proportion of isolates were resistant to antimicrobials, mainly quinolones and tetracycline, which are commonly used to treat campylobacteriosis in humans, emphasizing the importance of prudent antimicrobial use in veterinary medicine. This highlights the need for proper food safety and infection control practices to prevent the transmission of antimicrobial-resistant Campylobacter spp. through the consumption of contaminated poultry products.
Footnotes
Acknowledgments
All the authors express their appreciation to the University of Valencia (Department of Microbiology, Faculty of Pharmacy) and the Public Health Laboratory of Valencia, for the support that enabled this study to be prepared.
Authors' Contributions
B.B.: Protocol development, methodology, and writing—original draft preparation and editing. P.M.: Conceptualization, protocol development, and writing—review and editing. S.M.: Conceptualization, funding acquisition, protocol development, and writing—review and editing. M.M.: Protocol development, methodology, and writing—review and editing. M.I.: Conceptualization, funding acquisition, protocol development, and writing—original draft preparation, review, and editing.
Authorship Confirmation
All authors read and approved the final draft of the article and agreed to its publication.
Ethical Compliance
All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional and/or National Research Committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Disclosure Statement
No competing financial interests exist
Funding Information
This work was supported by the Spanish Ministry of Innovation and Science Grant AGL 2011-29382 (to S. M. and M.I.).
